- Email: [email protected]
Role of Fission Yeast Myosin I in Organization of Sterol-Rich Membrane Domains
Current Biology, Vol. 15, 1331–1336, July 26, 2005, ©2005 Elsevier Ltd All rights reserved. DOI 10.1016/j.cub.2005.07.009
Role of Fission Yeast Myosin I in Organization of Sterol-Rich Membrane Domains Tetsuya Takeda1 and Fred Chang* Department of Microbiology Columbia University College of Physicians and Surgeons New York, New York 10032
Summary Specialized membrane domains containing lipid rafts are thought to be important for membrane processes such as signaling and trafficking [1, 2]. An unconventional type I myosin has been shown to reside in lipid rafts and function to target a disaccharidase to rafts in brush borders of intestinal mammalian cells [3]. In the fission yeast Schizosaccharomyces pombe, distinct sterol-rich membrane domains are formed at the cell division site and sites of polarized cell growth at cell tips [4]. Here, we show that the sole S. pombe myosin I, myo1p, is required for proper organization of these membrane domains. myo1 mutants lacking the TH1 domain exhibit a uniform distribution of sterol-rich membranes all over the plasma membrane throughout the cell cycle. These effects are independent of endocytosis because myo1 mutants exhibit no endocytic defects. Conversely, overexpression of myo1p induces ectopic sterol-rich membrane domains. Myo1p localizes to nonmotile foci that cluster in sterol-rich plasma membrane domains and fractionates with detergent-resistant membranes. Because the myo1p TH1 domain may bind directly to acidic phospholipids, these findings suggest a model for how type I myosin contributes to the organization of specialized membrane domains. Results and Discussion S. pombe has one myosin I, myo1p, which is required for proper cell shape, actin organization, and cell-wall regulation but is not essential for cell viability [5, 6]. Myo1p is composed of a head (motor) domain and TH1, TH2, TH3, and A domains in its tail region (Figure 1A). Based on similarity to other myosin Is [7, 8], the basic TH1 domain is thought to bind to acidic phospholipids. Other regions of the myo1p tail (including the acidic A domain) stimulate the actin nucleation activity of Arp2/ 3 complex in a functionally redundant manner with WASp [5, 9, 10]. Specialized sterol-rich membrane domains at sites of cell division and cell growth can be visualized in fission yeast by staining with filipin, a fluorescent sterol binding dye [4]. Numerous tests have shown that filipin can be used as a reliable marker for these membrane domains in fission yeast [11]. Another marker for mem*Correspondence: [email protected] 1 Present address: Cancer Research UK Cell Cycle Genetics Research Group, Department of Genetics, University of Cambridge, Cambridge CB2 3EH, United Kingdom.
brane domains in these cells is an acylated GFP fusion (Ac-GFP) [11, 12]. We found that fission yeast myo1D mutants lack distinct membrane domains. Filipin staining in wild-type cells shows increased fluorescence intensity in distinct membrane domain at cell tips and at the septum (Figure 1B) [4]. In contrast, myo1D cells showed a uniform distribution of filipin staining in the plasma membrane throughout the cell cycle (Figure 1B). Fluorescence intensity measurements showed that differences were significant: in myo1D cells, the membrane on the sides, cell tips, and septum were nearly equivalent, at levels normally seen at cell tips and septum in wild-type cells (Figure 1B). Ac-GFP also showed abnormal distribution in the myo1D mutant. In wild-type cells, a distinct region of increased Ac-GFP fluorescence is normally formed in a medial membrane domain at the future division site beginning in anaphase (although its localization less distinct at cell tips) [11]. Consistent with the filipin staining, a distinct medial Ac-GFP band was not visible in myo1D cells during anaphase (Figure 1C). To determine what regions of the myo1p protein are needed for membrane-domain organization, we generated C-terminal truncation alleles of myo1 in the chromosomal myo1 locus (Figure 1A). Consistent with previous work with plasmid-based constructs [5], the phenotypes of these mutants showed that the TH1 domain, but not the other tail domains, was critical for the functions of myo1p in F-actin organization, efficient septation, polarized cell growth, and viability under stress conditions (data not shown). Filipin staining showed that the TH2, TH3, and A domains were not required for proper membrane organization (Figure 1B, myo1(H1)). However, a truncation of the whole-tail region (the TH1, TH2, TH3, and A domains) resembled (and produced even slightly sicker phenotype than) the myo1 deletion (Figure 1B, myo1(H)). Ac-GFP localization corroborated the filipin staining patterns in these mutants (data not shown). Thus, the myo1p TH1 domain, which contains a putative phospholipid binding site, is necessary for membrane organization. We considered whether these effects of myo1 deletion are secondary to its possible effects on actin distribution, cell polarity, or endocytosis. Earlier studies with latrunculin A-treated cells have showed that F-actin is not required for formation of the medial membrane domain [4, 11]. In addition, cdc3 (profilin) mutant cells, which exhibit abnormal organization of all actin structures, still form normal filipin-staining patterns (Supplemental Data). for3 (formin) mutants lack actin cables and have defects in cell polarity [13]. Distinct membrane domains were also found for3D cells, even in the subset of those that are almost entirely round (Supplemental Data). Thus, it is unlikely that simply defects in cell polarization or actin distribution cause the membrane organization defects seen in myo1 mutants. Many myosin Is have been implicated in endocytosis and are localized at sites of endocytosis in different cell types [14]. For instance, in budding yeast, two myosin Is (S. cerevisiae myo3p and myo5p) are necessary for
Current Biology 1332
Figure 1. Fission Yeast myo1 Mutants Are Defective in the Membrane-Domain Organization (A) Domain structure of myo1p and myo1p truncation mutants. myo1 tail truncation mutants were constructed by integrating a kanMX cassette into specific locations in the chromosomal myo1 locus. (B) Localization of sterol-rich membrane domains in myo1 mutants. Cells of the indicated genotype were grown at 25°C and stained for sterol-rich membrane domains with filipin. Graph shows quantitation of filipin staining. Maximum fluorescence intensities were measured at different points on the cell surface, and relative intensities were determined with the intensities of septum in wild-type cells for septating cells and the intensities at cell tip in wild-type cells for interphase cells. For direct comparison of fluorescent intensities, wild-type cells marked with an additional GFP marker were mixed with the mutants and imaged in the same field. Error bars represent the standard error of the mean. (C) Wild-type and myo1-null cells expressing Ac-GFP (a lipid-raft marker) were imaged. Hoechst 3342 staining show that the cells are in anaphase. Arrowheads indicate a band of increased Ac-GFP fluorescence at the future division site in wild-type cells. Bar, 10 m.
Myosin I and Membrane Domains 1333
Figure 2. Myo1p Is Not Required for Endocytosis (A) FM4-64-internalization assay for endocytosis. Cells were labeled with 20 M FM4-64 on ice for 15 min, washed with ice-cold medium, and resuspended in fresh YE5S medium and then incubated at 30°C for the indicated time period. (B) Luicfer yellow uptake assay for fluid-phase endocytosis. Cells were incubated with 4 mg/ml lucifer yellow for 30 min at 30°C and imaged.
efficient endocytosis and are thought to contribute to scission of endocytic vesicles from the plasma membrane [15–17]. We can imagine that endocytosis could affect membrane composition by removing excess sterol-rich membranes from the cell surface and that myo1p could have a primary function in endocytosis. However, we found that myo1p is not essential for endocytosis in fission yeast. We used standard endocytosis assays measuring the uptake of a lipid dye FM464, which assays membrane uptake, and lucifer yellow, which assays fluid phase uptake [18, 19]. No defects in myo1D cells were seen in the dynamics of FM4-46 or lucifer yellow entry into the cell and trafficking to vacuoles in cells after recovery from cold treatment (Figure 2). In addition, we also detected no delays in FM4-64 internalization in cells without cold treatment, as assayed in time-lapse imaging of early time points (5–20 min after FM4-64 addition) (data not shown). These data do not rule out the possibility that myo1p contributes to endocytosis but is functionally redundant with other proteins, such as WASp [5]. The results do say however that the defects in membrane organization in myo1 mutants are not simply due to endocytic defects. Further, latrunculin A-treated cells, which are strongly defective in endocytosis, still form membrane domains, demonstrating that endocytosis per se is not required for this process [4, 11]. The overexpression of myo1p led to formation of ectopic membrane domains. We overexpressed a functional GFP-myo1p from the strong nmt1 promoter in cdc25 temperature-sensitive cells arrested in G2 phase. Normally, G2-arrested cells exhibit no filipinstaining domains on the cell sides (Figures 3A and 3B, uninduced). GFP-myo1p overexpression caused the formation of abnormal medial membrane domains on the sides of most cells (Figures 3A and 3B, induced) (68.6%, n = 156 versus 7.4%, n = 135 in uninduced cells). GFP-myo1p dots were clustered at these abnormal membrane domains (Figure 3C). DAPI staining con-
firmed that cells were arrested in G2-phase because the large majority of these cells (>91%) were mononucleate (data not shown). It was not possible to assay effects of overexpressing the TH1 domain alone because this protein fragment targets abnormally to the nucleus [5]. We next tested whether myo1p associates with sterol-rich membrane domains. A myo1p-GFP fusion localized to discrete foci that concentrate at sites of cell growth at cell tips and septum and also at the contractile ring (Figure 4A, wild-type; see also [5]). The GFP fusion was considered functional because the cells expressing the fusion as its only myo1p exhibited no obvious abnormalities in cell shape or membrane staining. Concurrent staining with filipin revealed that these myo1p-GFP foci resided largely in sterol-rich membrane domains (93.2% of foci, n = 148; Figure 4A). Recently, we have found that membrane domains are often organized in spiral patterns in cdc15 (PCH family protein) cytokinesis mutants [11]. Strikingly, myo1pGFP foci were clustered in these spiral-shaped membrane domains (Figure 4A, cdc15-287). Overexpression of cdc15p often causes formation of ectopic membrane domains [11], accumulation of myo1p-GFP in the medial region [20], and much less often, actin rings [21]. Concurrent staining with filipin showed that myo1pGFP foci clustered in these ectopic membrane domains (Figure 4A, cdc15 OE cdc25-22) (100% of ectopic membrane domains have clusters of myo1-GFP, n = 18). To test whether the localization of myo1-GFP is actin dependent, we examined myo1-GFP in cells treated with Lat-A for 5–10 min, which leads to complete disassembly of the actin cytoskeleton. In contrast to a previous report [5], using confocal microscopy, we detected small amounts of myo1-GFP associated with the plasma membrane in the region of the sterol-rich membrane domains at the cell tips, medial band, and septa (Figure 4B). Thus, myo1-GFP can associate in vivo with sterol-rich membranes in an actin-independent manner.
Current Biology 1334
Figure 3. Myo1p Overexpression Induces Ectopic Membrane Domains (A) cdc25-22 cells with pSGP573-myo1 were grown in EMM + thiamine media, washed, resuspended in EMM + thiamine (uninduced) or EMM without thiamine (induced) media, grown for 15 hr at 25°C, shifted to 36°C for 3 hr to arrest cells in G2 phase, stained with filipin, and imaged. Cells were confirmed to be in interphase by DAPI staining (data not shown). Bar, 10 m. (B) Graph shows percentage of cells with filipin staining only at cell tips (tips) or staining in medial regions and cell tips (medial). (C) GFP-myo1p fluorescence and filipin staining in cells with ectopic membrane domains, as described above. Arrows point to clusters of GFP-myo1p at ectopic regions of sterol-rich membrane domains.
A biochemical test commonly used for lipid raft components is their ability to fractionate with detergentresistant membranes (DRMs). Yeast extracts expressing a functional myo1p-HA fusion were treated with cold Triton X-100 and fractionated by step sucrose gradients. Myo1p-HA was present in all fractions, and 10% of the total appeared in the top floating “lipid-raft” fraction, marked by the lipid-raft marker pma1p (Figure 4C, first gradient, fraction 1). To further test for lipid-raft association, we reloaded the top fraction from the first gradient onto a second gradient and found that a significant portion of myo1p again associated with the top lipid-raft fraction after a second centrifugation (Figure 4C, second gradient, fraction 1). Some myo1p sedimented as heavier particles even in the second gradient; we speculate that these may have lost membrane association or associated with other proteins such as actin during the procedure. The Ac-GFP protein also cofractioned in the lipid-raft fraction, showing a similar behavior to myo1p in these types of gradients (data not shown; see also [12]). Thus, both imaging and biochemical analysis showed that myo1p is a component of sterol-rich membrane domains. The fact that myo1p is a putative motor protein suggests a model in which myo1p organizes large mem-
brane domains by moving together smaller patches of sterol-rich plasma membranes to areas such as the cell-division site. To test this possibility, we imaged myo1p-GFP in living cells by time-lapse confocal microscopy. Myo1p-GFP foci generally did not move in the plane of membrane but flashed on and off transiently with an average duration of 12.8 ± 0.9 s (n = 37 foci) (Movie S1), very similar to myosin type I foci at actin patches in budding yeast (lifespan of 13 s) [16, 17]. During initial stages of medial sterol-rich membrane-domain formation in anaphase, myo1p-GFP foci were delocalized throughout the sides of the cells before concentrating at the septum (Figure 5A). Kymographic analyses of time-lapse sequences showed that during this period, the myo1p foci were also transient and did not exhibit significant lateral movements (Figure 5B). Imaging of myo1-GFP foci in the focal plane of the plasma membrane also did not reveal lateral movements. Thus, we found little evidence that myo1p functions to form membrane domains simply by generating lateral movements toward the future division site, although we cannot rule out that there may be a subset of myo1p proteins that are not apparent by our imaging. In summary, we have identified a role for S. pombe myosin I in the organization of membrane domains. Our
Myosin I and Membrane Domains 1335
Figure 4. Myo1p Associates with Sterol-Rich Membrane Domains (A) Images of myo1-GFP in wild-type, cdc15287 cells and interphase cells overexpressing cdc15+ stained with filipin. Cells overexpressing cdc15+ were myo1-GFP cdc25-22 cells with REP1-cdc15, which were grown in EMM + thiamine media, washed, resuspended in EMM + thiamine (uninduced) or EMM without thiamine (induced) media, grown for 15 hr at 25°C, shifted to 36°C for 3 hr, and then stained with filipin and imaged. Myo1-GFP foci localize in the membrane domains (arrowheads). Bar, 10 m. (B) Images of cells expressing myo1-GFP treated with latrunculin A for 5 min (top) or 10 min (bottom). Arrows point to residual staining of myo1-GFP at regions of the cell tips and septa. (C) Association of myo1-HA with detergentresistant membranes. Yeast extracts were treated with cold 1% Triton X-100 and loaded onto the bottom of a 30%–35% Optiprep gradient. After centrifugation, the top (lipid raft) two fractions from the first gradient were loaded onto a second Optiprep gradient and centrifuged again. Fractions from gradients were immunoblotted with anti-HA or anti-pma1p (a lipid-raft marker) antibodies or Coomassie stained for total proteins.
study extends on findings that a mammalian brush-border type I myosin (myosin-1A) is a lipid-raft protein that is required for the targeting of other proteins to lipid rafts and for normal brush-border morphology [3, 22]. Our findings show that fission yeast type I myosin has a global role in the general organization of sterol-rich membrane domains. It is not yet clear how myo1p regulates membrane composition and if its effects are direct. One model is that the basic TH1 domain of myo1p, which is essential for this myo1p function, binds to phospholipids that directly affect the local composition of the membrane. The clustering of acidic phospholipids by myo1p foci could plausibly exert long-range influence on membrane composition [23]. The TH1 domain could also be responsible for targeting of myo1p to the membrane. A second model is that myo1p may contribute to membrane tension, as shown in Dictyostelium [24], which could be imagined to affect membrane organization. However, actin depolymerization does not cause the same effects as myo1 disruption in fission yeast [4, 11]. It is likely that myo1p functions at sites of endocytosis (actin patches), although the effects of myo1p on membrane domains is separable
from endocytosis. An interaction between myo1p and the contractile ring protein cdc15p (PCH family protein), which also regulates membrane domains, suggests that myo1p functions together with other proteins [11, 20]. The precise function of these membrane domains is still not clear, although possible effects on membrane dynamics, signaling, and contractile-ring positioning have been proposed. From the phenotypes of myo1 mutants, we can begin to infer that the organization of the plasma membrane into distinct membrane domains is not essential for viability, polarized growth, or celldivision-site placement. However, these domains appear to affect spatial distribution on cell-wall dynamics and endocytic sites. Further study of the different functions of myosin I will provide insights into the coordination between actin assembly, endocytosis, and membrane domains at the plasma membrane. Supplemental Data Supplemental Data include Supplemental Experimental Procedures, one figure, one table, and one movie and can be found with this article online at http://www.current-biology.com/cgi/ content/full/15/14/1331/DC1/.
Current Biology 1336
Figure 5. Myo1p Does Not Move in the Plane of the Plasma Membrane (A) Redistribution of myo1p-GFP foci to the medial division site during mitosis. Time-lapse images every 5 min were taken in 10 Z planes 0.5 m apart with a spinning disc confocal. Maximum intensity projection images are shown. Locations of the dividing nuclei are outlined in blue. Bar, 10 m. (B) Dynamics of myo1-GFP foci during medial sterol-rich-domain formation. Time-lapse confocal images were taken every 1.8 s in a single focal plane. Kymograph of myo1-GFP on the side of the cell is presented. Bar, 10 m.
Acknowledgments We thank A. Chang, K. Gould, V. Sirotkin, and T. Pollard for valuable reagents, S. Salas-Pino and R. Daga for technical support, and members of our lab for discussion and advice. This work was supported by National Institutes of Health R01 GM056836 to F.C. and Postdoctoral Fellowships from Uehara Memorial Foundation to T.T. Received: March 24, 2005 Revised: May 27, 2005 Accepted: June 13, 2005 Published: July 26, 2005 References 1. Mukherjee, S., and Maxfield, F.R. (2004). Membrane domains. Annu. Rev. Cell Dev. Biol. 20, 839–866. 2. Rajendran, L., and Simons, K. (2005). Lipid rafts and membrane dynamics. J. Cell Sci. 118, 1099–1102. 3. Tyska, M.J., and Mooseker, M.S. (2004). A role for myosin-1A in the localization of a brush border disaccharidase. J. Cell Biol. 165, 395–405.
4. Wachtler, V., Rajagopalan, S., and Balasubramanian, M.K. (2003). Sterol-rich plasma membrane domains in the fission yeast Schizosaccharomyces pombe. J. Cell Sci. 116, 867–874. 5. Lee, W.L., Bezanilla, M., and Pollard, T.D. (2000). Fission yeast myosin-I, Myo1p, stimulates actin assembly by Arp2/3 complex and shares functions with WASp. J. Cell Biol. 151, 789– 800. 6. Toya, M., Motegi, F., Nakano, K., Mabuchi, I., and Yamamoto, M. (2001). Identification and functional analysis of the gene for type I myosin in fission yeast. Genes Cells 6, 187–199. 7. Adams, R.J., and Pollard, T.D. (1989). Binding of myosin I to membrane lipids. Nature 340, 565–568. 8. Doberstein, S.K., and Pollard, T.D. (1992). Localization and specificity of the phospholipid and actin binding sites on the tail of Acanthamoeba myosin IC. J. Cell Biol. 117, 1241–1249. 9. Lechler, T., Shevchenko, A., Shevchenko, A., and Li, R. (2000). Direct involvement of yeast type I myosins in Cdc42-dependent actin polymerization. J. Cell Biol. 148, 363–373. 10. Evangelista, M., Klebl, B.M., Tong, A.H., Webb, B.A., Leeuw, T., Leberer, E., Whiteway, M., Thomas, D.Y., and Boone, C. (2000). A role for myosin-I in actin assembly through interactions with Vrp1p, Bee1p, and the Arp2/3 complex. J. Cell Biol. 148, 353– 362. 11. Takeda, T., Kawate, T., and Chang, F. (2004). Organization of a sterol-rich membrane domain by cdc15p during cytokinesis in fission yeast. Nat. Cell Biol. 6, 1142–1144. 12. Zacharias, D.A., Violin, J.D., Newton, A.C., and Tsien, R.Y. (2002). Partitioning of lipid-modified monomeric GFPs into membrane microdomains of live cells. Science 296, 913–916. 13. Feierbach, B., and Chang, F. (2001). Roles of the fission yeast formin for3p in cell polarity, actin cable formation and symmetric cell division. Curr. Biol. 11, 1656–1665. 14. Engqvist-Goldstein, A.E., and Drubin, D.G. (2003). Actin assembly and endocytosis: from yeast to mammals. Annu. Rev. Cell Dev. Biol. 19, 287–332. 15. Geli, M.I., and Riezman, H. (1996). Role of type I myosins in receptor-mediated endocytosis in yeast. Science 272, 533– 535. 16. Goodson, H.V., Anderson, B.L., Warrick, H.M., Pon, L.A., and Spudich, J.A. (1996). Synthetic lethality screen identifies a novel yeast myosin I gene (MYO5): myosin I proteins are required for polarization of the actin cytoskeleton. J. Cell Biol. 133, 1277–1291. 17. Jonsdottir, G.A., and Li, R. (2004). Dynamics of yeast Myosin I: evidence for a possible role in scission of endocytic vesicles. Curr. Biol. 14, 1604–1609. 18. Vida, T.A., and Emr, S.D. (1995). A new vital stain for visualizing vacuolar membrane dynamics and endocytosis in yeast. J. Cell Biol. 128, 779–792. 19. Wiederkehr, A., Meier, K.D., and Riezman, H. (2001). Identification and characterization of Saccharomyces cerevisiae mutants defective in fluid-phase endocytosis. Yeast 18, 759–773. 20. Carnahan, R.H., and Gould, K.L. (2003). The PCH family protein, Cdc15p, recruits two F-actin nucleation pathways to coordinate cytokinetic actin ring formation in Schizosaccharomyces pombe. J. Cell Biol. 162, 851–862. 21. Fankhauser, C., Reymond, A., Cerutti, L., Utzig, S., Hofmann, K., and Simanis, V. (1995). The S. pombe cdc15 gene is a key element in the reorganization of F-actin at mitosis. Cell 82, 435–444. 22. Tyska, M.J., Mackey, A.T., Huang, J.D., Copland, N.G., Jenkins, N.A., and Mooseker, M.S. (2005). Myosin-1a is critical for normal brush border structure and composition. Mol. Biol. Cell 16, 2443–2457. 23. Papayannopoulos, V., Co, C., Prehoda, K.E., Snapper, S., Taunton, J., and Lim, W.A. (2005). A polybasic motif allows N-WASP to act as a sensor of PIP(2) density. Mol. Cell 17, 181–191. 24. Dai, J., Ting-Beall, H.P., Hochmuth, R.M., Sheetz, M.P., and Titus, M.A. (1999). Myosin I contributes to the generation of resting cortical tension. Biophys. J. 77, 1168–1176.
Role of Fission Yeast Myosin I in Organization of Sterol-Rich Membrane Domains Tetsuya Takeda1 and Fred Chang* Department of Microbiology Columbia University College of Physicians and Surgeons New York, New York 10032
Summary Specialized membrane domains containing lipid rafts are thought to be important for membrane processes such as signaling and trafficking [1, 2]. An unconventional type I myosin has been shown to reside in lipid rafts and function to target a disaccharidase to rafts in brush borders of intestinal mammalian cells [3]. In the fission yeast Schizosaccharomyces pombe, distinct sterol-rich membrane domains are formed at the cell division site and sites of polarized cell growth at cell tips [4]. Here, we show that the sole S. pombe myosin I, myo1p, is required for proper organization of these membrane domains. myo1 mutants lacking the TH1 domain exhibit a uniform distribution of sterol-rich membranes all over the plasma membrane throughout the cell cycle. These effects are independent of endocytosis because myo1 mutants exhibit no endocytic defects. Conversely, overexpression of myo1p induces ectopic sterol-rich membrane domains. Myo1p localizes to nonmotile foci that cluster in sterol-rich plasma membrane domains and fractionates with detergent-resistant membranes. Because the myo1p TH1 domain may bind directly to acidic phospholipids, these findings suggest a model for how type I myosin contributes to the organization of specialized membrane domains. Results and Discussion S. pombe has one myosin I, myo1p, which is required for proper cell shape, actin organization, and cell-wall regulation but is not essential for cell viability [5, 6]. Myo1p is composed of a head (motor) domain and TH1, TH2, TH3, and A domains in its tail region (Figure 1A). Based on similarity to other myosin Is [7, 8], the basic TH1 domain is thought to bind to acidic phospholipids. Other regions of the myo1p tail (including the acidic A domain) stimulate the actin nucleation activity of Arp2/ 3 complex in a functionally redundant manner with WASp [5, 9, 10]. Specialized sterol-rich membrane domains at sites of cell division and cell growth can be visualized in fission yeast by staining with filipin, a fluorescent sterol binding dye [4]. Numerous tests have shown that filipin can be used as a reliable marker for these membrane domains in fission yeast [11]. Another marker for mem*Correspondence: [email protected] 1 Present address: Cancer Research UK Cell Cycle Genetics Research Group, Department of Genetics, University of Cambridge, Cambridge CB2 3EH, United Kingdom.
brane domains in these cells is an acylated GFP fusion (Ac-GFP) [11, 12]. We found that fission yeast myo1D mutants lack distinct membrane domains. Filipin staining in wild-type cells shows increased fluorescence intensity in distinct membrane domain at cell tips and at the septum (Figure 1B) [4]. In contrast, myo1D cells showed a uniform distribution of filipin staining in the plasma membrane throughout the cell cycle (Figure 1B). Fluorescence intensity measurements showed that differences were significant: in myo1D cells, the membrane on the sides, cell tips, and septum were nearly equivalent, at levels normally seen at cell tips and septum in wild-type cells (Figure 1B). Ac-GFP also showed abnormal distribution in the myo1D mutant. In wild-type cells, a distinct region of increased Ac-GFP fluorescence is normally formed in a medial membrane domain at the future division site beginning in anaphase (although its localization less distinct at cell tips) [11]. Consistent with the filipin staining, a distinct medial Ac-GFP band was not visible in myo1D cells during anaphase (Figure 1C). To determine what regions of the myo1p protein are needed for membrane-domain organization, we generated C-terminal truncation alleles of myo1 in the chromosomal myo1 locus (Figure 1A). Consistent with previous work with plasmid-based constructs [5], the phenotypes of these mutants showed that the TH1 domain, but not the other tail domains, was critical for the functions of myo1p in F-actin organization, efficient septation, polarized cell growth, and viability under stress conditions (data not shown). Filipin staining showed that the TH2, TH3, and A domains were not required for proper membrane organization (Figure 1B, myo1(H1)). However, a truncation of the whole-tail region (the TH1, TH2, TH3, and A domains) resembled (and produced even slightly sicker phenotype than) the myo1 deletion (Figure 1B, myo1(H)). Ac-GFP localization corroborated the filipin staining patterns in these mutants (data not shown). Thus, the myo1p TH1 domain, which contains a putative phospholipid binding site, is necessary for membrane organization. We considered whether these effects of myo1 deletion are secondary to its possible effects on actin distribution, cell polarity, or endocytosis. Earlier studies with latrunculin A-treated cells have showed that F-actin is not required for formation of the medial membrane domain [4, 11]. In addition, cdc3 (profilin) mutant cells, which exhibit abnormal organization of all actin structures, still form normal filipin-staining patterns (Supplemental Data). for3 (formin) mutants lack actin cables and have defects in cell polarity [13]. Distinct membrane domains were also found for3D cells, even in the subset of those that are almost entirely round (Supplemental Data). Thus, it is unlikely that simply defects in cell polarization or actin distribution cause the membrane organization defects seen in myo1 mutants. Many myosin Is have been implicated in endocytosis and are localized at sites of endocytosis in different cell types [14]. For instance, in budding yeast, two myosin Is (S. cerevisiae myo3p and myo5p) are necessary for
Current Biology 1332
Figure 1. Fission Yeast myo1 Mutants Are Defective in the Membrane-Domain Organization (A) Domain structure of myo1p and myo1p truncation mutants. myo1 tail truncation mutants were constructed by integrating a kanMX cassette into specific locations in the chromosomal myo1 locus. (B) Localization of sterol-rich membrane domains in myo1 mutants. Cells of the indicated genotype were grown at 25°C and stained for sterol-rich membrane domains with filipin. Graph shows quantitation of filipin staining. Maximum fluorescence intensities were measured at different points on the cell surface, and relative intensities were determined with the intensities of septum in wild-type cells for septating cells and the intensities at cell tip in wild-type cells for interphase cells. For direct comparison of fluorescent intensities, wild-type cells marked with an additional GFP marker were mixed with the mutants and imaged in the same field. Error bars represent the standard error of the mean. (C) Wild-type and myo1-null cells expressing Ac-GFP (a lipid-raft marker) were imaged. Hoechst 3342 staining show that the cells are in anaphase. Arrowheads indicate a band of increased Ac-GFP fluorescence at the future division site in wild-type cells. Bar, 10 m.
Myosin I and Membrane Domains 1333
Figure 2. Myo1p Is Not Required for Endocytosis (A) FM4-64-internalization assay for endocytosis. Cells were labeled with 20 M FM4-64 on ice for 15 min, washed with ice-cold medium, and resuspended in fresh YE5S medium and then incubated at 30°C for the indicated time period. (B) Luicfer yellow uptake assay for fluid-phase endocytosis. Cells were incubated with 4 mg/ml lucifer yellow for 30 min at 30°C and imaged.
efficient endocytosis and are thought to contribute to scission of endocytic vesicles from the plasma membrane [15–17]. We can imagine that endocytosis could affect membrane composition by removing excess sterol-rich membranes from the cell surface and that myo1p could have a primary function in endocytosis. However, we found that myo1p is not essential for endocytosis in fission yeast. We used standard endocytosis assays measuring the uptake of a lipid dye FM464, which assays membrane uptake, and lucifer yellow, which assays fluid phase uptake [18, 19]. No defects in myo1D cells were seen in the dynamics of FM4-46 or lucifer yellow entry into the cell and trafficking to vacuoles in cells after recovery from cold treatment (Figure 2). In addition, we also detected no delays in FM4-64 internalization in cells without cold treatment, as assayed in time-lapse imaging of early time points (5–20 min after FM4-64 addition) (data not shown). These data do not rule out the possibility that myo1p contributes to endocytosis but is functionally redundant with other proteins, such as WASp [5]. The results do say however that the defects in membrane organization in myo1 mutants are not simply due to endocytic defects. Further, latrunculin A-treated cells, which are strongly defective in endocytosis, still form membrane domains, demonstrating that endocytosis per se is not required for this process [4, 11]. The overexpression of myo1p led to formation of ectopic membrane domains. We overexpressed a functional GFP-myo1p from the strong nmt1 promoter in cdc25 temperature-sensitive cells arrested in G2 phase. Normally, G2-arrested cells exhibit no filipinstaining domains on the cell sides (Figures 3A and 3B, uninduced). GFP-myo1p overexpression caused the formation of abnormal medial membrane domains on the sides of most cells (Figures 3A and 3B, induced) (68.6%, n = 156 versus 7.4%, n = 135 in uninduced cells). GFP-myo1p dots were clustered at these abnormal membrane domains (Figure 3C). DAPI staining con-
firmed that cells were arrested in G2-phase because the large majority of these cells (>91%) were mononucleate (data not shown). It was not possible to assay effects of overexpressing the TH1 domain alone because this protein fragment targets abnormally to the nucleus [5]. We next tested whether myo1p associates with sterol-rich membrane domains. A myo1p-GFP fusion localized to discrete foci that concentrate at sites of cell growth at cell tips and septum and also at the contractile ring (Figure 4A, wild-type; see also [5]). The GFP fusion was considered functional because the cells expressing the fusion as its only myo1p exhibited no obvious abnormalities in cell shape or membrane staining. Concurrent staining with filipin revealed that these myo1p-GFP foci resided largely in sterol-rich membrane domains (93.2% of foci, n = 148; Figure 4A). Recently, we have found that membrane domains are often organized in spiral patterns in cdc15 (PCH family protein) cytokinesis mutants [11]. Strikingly, myo1pGFP foci were clustered in these spiral-shaped membrane domains (Figure 4A, cdc15-287). Overexpression of cdc15p often causes formation of ectopic membrane domains [11], accumulation of myo1p-GFP in the medial region [20], and much less often, actin rings [21]. Concurrent staining with filipin showed that myo1pGFP foci clustered in these ectopic membrane domains (Figure 4A, cdc15 OE cdc25-22) (100% of ectopic membrane domains have clusters of myo1-GFP, n = 18). To test whether the localization of myo1-GFP is actin dependent, we examined myo1-GFP in cells treated with Lat-A for 5–10 min, which leads to complete disassembly of the actin cytoskeleton. In contrast to a previous report [5], using confocal microscopy, we detected small amounts of myo1-GFP associated with the plasma membrane in the region of the sterol-rich membrane domains at the cell tips, medial band, and septa (Figure 4B). Thus, myo1-GFP can associate in vivo with sterol-rich membranes in an actin-independent manner.
Current Biology 1334
Figure 3. Myo1p Overexpression Induces Ectopic Membrane Domains (A) cdc25-22 cells with pSGP573-myo1 were grown in EMM + thiamine media, washed, resuspended in EMM + thiamine (uninduced) or EMM without thiamine (induced) media, grown for 15 hr at 25°C, shifted to 36°C for 3 hr to arrest cells in G2 phase, stained with filipin, and imaged. Cells were confirmed to be in interphase by DAPI staining (data not shown). Bar, 10 m. (B) Graph shows percentage of cells with filipin staining only at cell tips (tips) or staining in medial regions and cell tips (medial). (C) GFP-myo1p fluorescence and filipin staining in cells with ectopic membrane domains, as described above. Arrows point to clusters of GFP-myo1p at ectopic regions of sterol-rich membrane domains.
A biochemical test commonly used for lipid raft components is their ability to fractionate with detergentresistant membranes (DRMs). Yeast extracts expressing a functional myo1p-HA fusion were treated with cold Triton X-100 and fractionated by step sucrose gradients. Myo1p-HA was present in all fractions, and 10% of the total appeared in the top floating “lipid-raft” fraction, marked by the lipid-raft marker pma1p (Figure 4C, first gradient, fraction 1). To further test for lipid-raft association, we reloaded the top fraction from the first gradient onto a second gradient and found that a significant portion of myo1p again associated with the top lipid-raft fraction after a second centrifugation (Figure 4C, second gradient, fraction 1). Some myo1p sedimented as heavier particles even in the second gradient; we speculate that these may have lost membrane association or associated with other proteins such as actin during the procedure. The Ac-GFP protein also cofractioned in the lipid-raft fraction, showing a similar behavior to myo1p in these types of gradients (data not shown; see also [12]). Thus, both imaging and biochemical analysis showed that myo1p is a component of sterol-rich membrane domains. The fact that myo1p is a putative motor protein suggests a model in which myo1p organizes large mem-
brane domains by moving together smaller patches of sterol-rich plasma membranes to areas such as the cell-division site. To test this possibility, we imaged myo1p-GFP in living cells by time-lapse confocal microscopy. Myo1p-GFP foci generally did not move in the plane of membrane but flashed on and off transiently with an average duration of 12.8 ± 0.9 s (n = 37 foci) (Movie S1), very similar to myosin type I foci at actin patches in budding yeast (lifespan of 13 s) [16, 17]. During initial stages of medial sterol-rich membrane-domain formation in anaphase, myo1p-GFP foci were delocalized throughout the sides of the cells before concentrating at the septum (Figure 5A). Kymographic analyses of time-lapse sequences showed that during this period, the myo1p foci were also transient and did not exhibit significant lateral movements (Figure 5B). Imaging of myo1-GFP foci in the focal plane of the plasma membrane also did not reveal lateral movements. Thus, we found little evidence that myo1p functions to form membrane domains simply by generating lateral movements toward the future division site, although we cannot rule out that there may be a subset of myo1p proteins that are not apparent by our imaging. In summary, we have identified a role for S. pombe myosin I in the organization of membrane domains. Our
Myosin I and Membrane Domains 1335
Figure 4. Myo1p Associates with Sterol-Rich Membrane Domains (A) Images of myo1-GFP in wild-type, cdc15287 cells and interphase cells overexpressing cdc15+ stained with filipin. Cells overexpressing cdc15+ were myo1-GFP cdc25-22 cells with REP1-cdc15, which were grown in EMM + thiamine media, washed, resuspended in EMM + thiamine (uninduced) or EMM without thiamine (induced) media, grown for 15 hr at 25°C, shifted to 36°C for 3 hr, and then stained with filipin and imaged. Myo1-GFP foci localize in the membrane domains (arrowheads). Bar, 10 m. (B) Images of cells expressing myo1-GFP treated with latrunculin A for 5 min (top) or 10 min (bottom). Arrows point to residual staining of myo1-GFP at regions of the cell tips and septa. (C) Association of myo1-HA with detergentresistant membranes. Yeast extracts were treated with cold 1% Triton X-100 and loaded onto the bottom of a 30%–35% Optiprep gradient. After centrifugation, the top (lipid raft) two fractions from the first gradient were loaded onto a second Optiprep gradient and centrifuged again. Fractions from gradients were immunoblotted with anti-HA or anti-pma1p (a lipid-raft marker) antibodies or Coomassie stained for total proteins.
study extends on findings that a mammalian brush-border type I myosin (myosin-1A) is a lipid-raft protein that is required for the targeting of other proteins to lipid rafts and for normal brush-border morphology [3, 22]. Our findings show that fission yeast type I myosin has a global role in the general organization of sterol-rich membrane domains. It is not yet clear how myo1p regulates membrane composition and if its effects are direct. One model is that the basic TH1 domain of myo1p, which is essential for this myo1p function, binds to phospholipids that directly affect the local composition of the membrane. The clustering of acidic phospholipids by myo1p foci could plausibly exert long-range influence on membrane composition [23]. The TH1 domain could also be responsible for targeting of myo1p to the membrane. A second model is that myo1p may contribute to membrane tension, as shown in Dictyostelium [24], which could be imagined to affect membrane organization. However, actin depolymerization does not cause the same effects as myo1 disruption in fission yeast [4, 11]. It is likely that myo1p functions at sites of endocytosis (actin patches), although the effects of myo1p on membrane domains is separable
from endocytosis. An interaction between myo1p and the contractile ring protein cdc15p (PCH family protein), which also regulates membrane domains, suggests that myo1p functions together with other proteins [11, 20]. The precise function of these membrane domains is still not clear, although possible effects on membrane dynamics, signaling, and contractile-ring positioning have been proposed. From the phenotypes of myo1 mutants, we can begin to infer that the organization of the plasma membrane into distinct membrane domains is not essential for viability, polarized growth, or celldivision-site placement. However, these domains appear to affect spatial distribution on cell-wall dynamics and endocytic sites. Further study of the different functions of myosin I will provide insights into the coordination between actin assembly, endocytosis, and membrane domains at the plasma membrane. Supplemental Data Supplemental Data include Supplemental Experimental Procedures, one figure, one table, and one movie and can be found with this article online at http://www.current-biology.com/cgi/ content/full/15/14/1331/DC1/.
Current Biology 1336
Figure 5. Myo1p Does Not Move in the Plane of the Plasma Membrane (A) Redistribution of myo1p-GFP foci to the medial division site during mitosis. Time-lapse images every 5 min were taken in 10 Z planes 0.5 m apart with a spinning disc confocal. Maximum intensity projection images are shown. Locations of the dividing nuclei are outlined in blue. Bar, 10 m. (B) Dynamics of myo1-GFP foci during medial sterol-rich-domain formation. Time-lapse confocal images were taken every 1.8 s in a single focal plane. Kymograph of myo1-GFP on the side of the cell is presented. Bar, 10 m.
Acknowledgments We thank A. Chang, K. Gould, V. Sirotkin, and T. Pollard for valuable reagents, S. Salas-Pino and R. Daga for technical support, and members of our lab for discussion and advice. This work was supported by National Institutes of Health R01 GM056836 to F.C. and Postdoctoral Fellowships from Uehara Memorial Foundation to T.T. Received: March 24, 2005 Revised: May 27, 2005 Accepted: June 13, 2005 Published: July 26, 2005 References 1. Mukherjee, S., and Maxfield, F.R. (2004). Membrane domains. Annu. Rev. Cell Dev. Biol. 20, 839–866. 2. Rajendran, L., and Simons, K. (2005). Lipid rafts and membrane dynamics. J. Cell Sci. 118, 1099–1102. 3. Tyska, M.J., and Mooseker, M.S. (2004). A role for myosin-1A in the localization of a brush border disaccharidase. J. Cell Biol. 165, 395–405.
4. Wachtler, V., Rajagopalan, S., and Balasubramanian, M.K. (2003). Sterol-rich plasma membrane domains in the fission yeast Schizosaccharomyces pombe. J. Cell Sci. 116, 867–874. 5. Lee, W.L., Bezanilla, M., and Pollard, T.D. (2000). Fission yeast myosin-I, Myo1p, stimulates actin assembly by Arp2/3 complex and shares functions with WASp. J. Cell Biol. 151, 789– 800. 6. Toya, M., Motegi, F., Nakano, K., Mabuchi, I., and Yamamoto, M. (2001). Identification and functional analysis of the gene for type I myosin in fission yeast. Genes Cells 6, 187–199. 7. Adams, R.J., and Pollard, T.D. (1989). Binding of myosin I to membrane lipids. Nature 340, 565–568. 8. Doberstein, S.K., and Pollard, T.D. (1992). Localization and specificity of the phospholipid and actin binding sites on the tail of Acanthamoeba myosin IC. J. Cell Biol. 117, 1241–1249. 9. Lechler, T., Shevchenko, A., Shevchenko, A., and Li, R. (2000). Direct involvement of yeast type I myosins in Cdc42-dependent actin polymerization. J. Cell Biol. 148, 363–373. 10. Evangelista, M., Klebl, B.M., Tong, A.H., Webb, B.A., Leeuw, T., Leberer, E., Whiteway, M., Thomas, D.Y., and Boone, C. (2000). A role for myosin-I in actin assembly through interactions with Vrp1p, Bee1p, and the Arp2/3 complex. J. Cell Biol. 148, 353– 362. 11. Takeda, T., Kawate, T., and Chang, F. (2004). Organization of a sterol-rich membrane domain by cdc15p during cytokinesis in fission yeast. Nat. Cell Biol. 6, 1142–1144. 12. Zacharias, D.A., Violin, J.D., Newton, A.C., and Tsien, R.Y. (2002). Partitioning of lipid-modified monomeric GFPs into membrane microdomains of live cells. Science 296, 913–916. 13. Feierbach, B., and Chang, F. (2001). Roles of the fission yeast formin for3p in cell polarity, actin cable formation and symmetric cell division. Curr. Biol. 11, 1656–1665. 14. Engqvist-Goldstein, A.E., and Drubin, D.G. (2003). Actin assembly and endocytosis: from yeast to mammals. Annu. Rev. Cell Dev. Biol. 19, 287–332. 15. Geli, M.I., and Riezman, H. (1996). Role of type I myosins in receptor-mediated endocytosis in yeast. Science 272, 533– 535. 16. Goodson, H.V., Anderson, B.L., Warrick, H.M., Pon, L.A., and Spudich, J.A. (1996). Synthetic lethality screen identifies a novel yeast myosin I gene (MYO5): myosin I proteins are required for polarization of the actin cytoskeleton. J. Cell Biol. 133, 1277–1291. 17. Jonsdottir, G.A., and Li, R. (2004). Dynamics of yeast Myosin I: evidence for a possible role in scission of endocytic vesicles. Curr. Biol. 14, 1604–1609. 18. Vida, T.A., and Emr, S.D. (1995). A new vital stain for visualizing vacuolar membrane dynamics and endocytosis in yeast. J. Cell Biol. 128, 779–792. 19. Wiederkehr, A., Meier, K.D., and Riezman, H. (2001). Identification and characterization of Saccharomyces cerevisiae mutants defective in fluid-phase endocytosis. Yeast 18, 759–773. 20. Carnahan, R.H., and Gould, K.L. (2003). The PCH family protein, Cdc15p, recruits two F-actin nucleation pathways to coordinate cytokinetic actin ring formation in Schizosaccharomyces pombe. J. Cell Biol. 162, 851–862. 21. Fankhauser, C., Reymond, A., Cerutti, L., Utzig, S., Hofmann, K., and Simanis, V. (1995). The S. pombe cdc15 gene is a key element in the reorganization of F-actin at mitosis. Cell 82, 435–444. 22. Tyska, M.J., Mackey, A.T., Huang, J.D., Copland, N.G., Jenkins, N.A., and Mooseker, M.S. (2005). Myosin-1a is critical for normal brush border structure and composition. Mol. Biol. Cell 16, 2443–2457. 23. Papayannopoulos, V., Co, C., Prehoda, K.E., Snapper, S., Taunton, J., and Lim, W.A. (2005). A polybasic motif allows N-WASP to act as a sensor of PIP(2) density. Mol. Cell 17, 181–191. 24. Dai, J., Ting-Beall, H.P., Hochmuth, R.M., Sheetz, M.P., and Titus, M.A. (1999). Myosin I contributes to the generation of resting cortical tension. Biophys. J. 77, 1168–1176.